atw 2018-10

atw Vol. 63 (2018) | Issue 10 ı October
| | Fig. 1.
One of the 5 fuel segments with a spring covered by Zr cladding at the top.
which covered the spring. The experimental
rod was irradiated in the
middle of a subassembly between
1985 and 1989 in the Goesgen PWR.
The Goesgen subassembly is a
15 X 15 fuel rod configuration with
20 water locations as can be seen in
Figure 2. The burn-up measured
segment was irradiated for 1226 EFPD
(Effective Full Power Days). The
average burn-up was 50.4 GWd/tHM.
The average linear power of the
segment was 260 W/cm [3].
| | Fig. 2.
The Goesgen Subassembly. The 4 bright blue
fuel elements in the middle depict equivalent
positions of the investigated fuel segment.
The water holes are seen as to some extent
enlarged red cycles within the dark blue color
of the simulated water. The green and yellow
pins are the rest fuel pins of the subassembly.
The equivalent subassembly that
was simulated is depicted in Figure 2.
The underlying numerical concept
was to meet the conditions as shown
above to which the fuel segment is
exposed to. In order to avoid statistical
error and without violating the above
mentioned constraints, 4 equivalent
fuel segments were simulated in the
center of the subassembly as can be
seen in Figure 2.
In Figure 2 the chosen option is
shown, namely the yellow pins were
considered as generating the power
which exhibits the flux and the power
defined for the fuel segments under
investigation. The green pins in
Figure 2 are also fuel pins, however
they were only considered as far as
their neutron generation was concerned.
This kind of simulation leads
to the requested linear power and
total burn-up level for the 1226 EFPD
irradiation period.
The flux solver MCNPX-CINDER
was used for the burn-up calculation.
In this way the reaction rate data are
transferred directly from the transport
solver to the burn-up module.
The burn up simulation of the
subassembly, shown in Figure 2, was
optimized to 5 steps. At the end of the
1226 EFPD a decay period of 9,600
days was considered. This is the period
between the end of the irradiation at
1,989 and the measurements of the
sample which were recently performed.
The radiochemical measurement
technique is generally based on alpha
spectrometry. α-particles, emitted
from the analyzed radionuclide, transfer
their energy to a semicon ductor,
thus creating electron-hole pairs. The
electron can be then detected to
account for the existence of the specific
radioactive nuclei under investigation.
The measurements in this work were
performed using the Canberra S100
(Canberra Industries Inc.) detector.
For the MA isotopes with characteristic
γ-rays e.g., Am-241, the
quantification was performed in
acidic digestion liquors obtained
from dissolution experiments performed
with irradiated Zry-4 using
γ-spec troscopy. The measurements
were performed by means of an extended
range coaxial Ge detector
(GX3018, Canberra Industries Inc.).
Table 1 presents the reference
results of the current study. It shows
the comparison between the cal culated
nuclides and combined measured isotopes
which have close (α spectrometry
based) peaks. The ratio of those results
are considered to be a reference
MAs 1) Pu-239 +
Pu-240
2) Pu-238 +
Am-241
as they exhibit the less statistical error
which is con sidered to be below 5% for
each α or γ type measurements. In particular
the ratio of the second and third
column is more reliable as the signal of
counting is higher. Based on Table 1
one can see that the comparison of calculated
and measured ratios between
the pairs “2” and “3” is very good. Using
the ratio of “1” leads to difference
of about 25 %. Further, the absolute
counting shows a factor 50 for columns
2 and 3 and factor 65 for column 1.
In order to extract more information
further measurements were
performed based on alpha spectrometry
for several Pu isotopes
and γ-spectroscopy for Am-241 as
mentioned above. Table 2 shows the
obtained results.
Analyzing the results in Tables 1
and 2 and accounting for the statistical
errors it can be seen that except
for the Measurement of Pu-240 all
results are consistent in a sense that
the contribution of the inner surface
of the cladding – which could not be
accounted for by the calculations- is
by about factor 50 higher than the
traces within the material. The larger
deviation due to the Pu-240 is consistent
in both tables, suggesting that
measurement of Pu-240 was counting
signals which did not come from
Pu-240 by about 40 %.
Conclusions
A burn-up simulation of a fuel rod
sample, which was irradiated in the
PWR Goesgen for 1,226 Effective Full
Power Days, was performed. In this
study the activity of several MA in the
Zry-4 cladding, covering the spring on
the tops of the sample, was measured
3) Cm-243 +
Cm-244
Ratio 2/3
Experimental 3.06 × 10 4 2.29 × 10 5 2.61 × 10 5 0.877
Calculated 4.57 × 10 2 4.52 × 10 3 5.26 × 10 3 0.859
| | Tab. 1.
Comparison between the measured and the calculated activities of 3 pairs of MA isotopes within the
Zircaloy of the fuel segment under investigation in unit Bq/(g Zry-4).
MAs Measured Calculated Ratio
measured/calculated
Pu-239 1.2 × 10 4 2.38 × 10 2 50.4
Pu-240 2.04 × 10 4 2.19 × 10 2 93
Pu-241 2.06 × 10 6 4.4 × 10 4 47
Pu-242 2.19 × 10 2 4.1 × 10 0 53
Am-241 1.72 × 10 5 3.76 × 10 3 46
| | Tab. 2.
Comparison between the measured and calculated activities of isotopes, using alpha spectrometry
for several Pu isotopes and γ-spectroscopy for Am241 within the Zircaloy of the fuel segment under
investigation in unit Bq/(g Zry-4).
DECOMMISSIONING AND WASTE MANAGEMENT 527
Decommissioning and Waste Management
Determination of Minor Actinides in Irradiated Fuel Rod Components ı Ron Dagan, Michel Herm, Volker Metz and Maarten Becker